Por favor, use este identificador para citar o enlazar este ítem:
http://conacyt.repositorioinstitucional.mx/jspui/handle/1000/7493| Prediction of COVID-19 Mortality to Support Patient Prognosis and Triage and Limits of Current Open-Source Data | |
| Riccardo Doyle | |
| Acceso Abierto | |
| Atribución-NoComercial-SinDerivadas | |
| https://doi.org/10.1101/2021.03.21.21253984 | |
| https://www.medrxiv.org/content/10.1101/2021.03.21.21253984v2 | |
| This study examines the accuracy and applicability of machine learning methods in early prediction of mortality in COVID-19 patients. Patient symptoms, pre-existing conditions, age and sex were employed as predictive attributes from data spanning 17 countries. Performance on a semi-evenly balanced class sample of 212 patients resulted in high detection accuracy of 92.5%, with strong specificity and sensitivity. Performance on a larger sample of 5,121 patients with only age and mortality information was added as a measure of baseline discriminatory ability. Stratifying - Random Forest - and linear - Logistic Regression - methods were applied, both achieving modestly strong performance, with 77.4%-79.3% sensitivity and 71.4%-72.6% accuracy, highlighting predictive power even on the basis of a single attribute. Mutual information was employed as a dimensionality reduction technique, greatly improving performance and showing how a small number of easily retrievable attributes can provide timely and accurate predictions, with applications for datasets with slowly available variables - such as laboratory results. Unlike existing studies making use of the same dataset, limitations of the data were extensively explored and detailed, as each results section outlines the main shortcomings of relevant analysis. Future use of this dataset should be cautious and always accompanied by disclaimers on issues of real-life reproducibility. While its open-source nature is a credit to the wider research community and more such datasets should be published, in its current state it can produce valid conclusions only for a limited set of applications, some of which were explored in this study. | |
| medRxiv and bioRxiv | |
| 02-04-2021 | |
| Preimpreso | |
| www.medrxiv.org | |
| Inglés | |
| Epidemia COVID-19 | |
| Público en general | |
| VIRUS RESPIRATORIOS | |
| Versión publicada | |
| publishedVersion - Versión publicada | |
| Aparece en las colecciones: | Artículos científicos |
Cargar archivos:
| Fichero | Tamaño | Formato | |
|---|---|---|---|
| Prediction of COVID 19 Mortality to support Patient Prognosis and Triage and Limits of Current Open Source Data.pdf | 505.55 kB | Adobe PDF | Visualizar/Abrir |